Another one bites the dust - My 2014 to 2021 ride has ended

About this time of year, 7 years ago, I began my journey with ST. I was thrilled to fully automate my house and kept looking for new opportunities to do so. The ST IDE said I had close to 300 devices (includes virtual, child devices, integrations, etc.) at its peak and a number of you probably know the pain that comes with it. Many posts from this community inspired me and helped me overcome the litany of issues that most hardcore ST users are way too familiar with. Well… I just got to the point I have had enough. I don’t know whether switching away will help solve my issues but doing nothing certainly will not.

I reached a breaking point several times during these past 7 years, and also purchased a Hubitat Elevation in hopes it would be a good replacement, or at the very least, take over all the automations that needed to be more timely & reliable (no cloud !!!). The reality is I never trusted it as a long term solution, and it is far from polished. I am however amazed at the work and constant updates the HE (apparently) small team is pushing out. They probably have a bright future especially if they can make it easier to use, polish it up a bit, and stay ‘local first’. I would expect more ST power users to switch to it just to untangle from the unreliable and slow cloud. Alas Hubitat was not the solution I was looking for…

The past few days I have excluded all my devices from ST and built up my zwave and zigbee mesh networks on Home Assistant. The hardware I picked to run it is the following:

Raspberry Pi 4 (4GB RAM)
Argon One M.2 SATA Case
deCONZ ConBee II (Zigbee 3.0 Controller)
Aeotec Z-Stick 7 (Zwave Series 700 Controller)

For those wondering, the RPI boots directly to the SATA SSD drive so I have no SD Card on my RPI4. This will allow for much faster speeds and better reliability (SD Cards are pretty bad choice for high read/write systems). The size of the SSD is simply because it was cheap otherwise they recommend using 128GB.

I have to say that the setup process was easier than I thought it would be, but certainly not ST easy. I like to say that ST is being “Applefied” meaning that they are trying to turn it into a ‘one button toaster’, so to speak, which anyone can operate regardless of their expertise. I do NOT like that trend as this typically comes with fewer options, and a tighter grip on the ecosystem by the company developing it. In other words, the users trade off control, flexibility and options for a more reliable system that doesn’t require constant work. Not sure this is a thorough description but I’d say ST is certainly not fully “Applefied” as for now I only see less control, fewer options and certainly not an equivalent reliability level. Maybe after this big evolution they are going through…

Why now? For a while my HomeSeer (never again!) dimmers & switches have been freezing requiring a power cycle. This is apparently due to the series 500 processor choking up on excessive data on my Zwave mesh network (BTW, @Kianoosh_Karami thanks for all the times you have helped me!!). Recently I found a few of these devices excluded even though I hadn’t done it myself… The big surprise came when I could not add it back as I was over 200 devices! The previous Android app still allowed it, but the latest update closed the “loop hole”. That was the last straw that broke the camel’s back… Obviously ST doesn’t want users with a high number of devices & integrations so I am no longer welcome and I quit.

What have I found these past few days on Home Assistant? SPEEEEEEEEEEEED ! Whoa I am impressed at how fast stuff happens, but it is not a surprise given there is no need for the cloud (love it!)! There are things that require the cloud, such as Alexa integration, but device control is all local. All my devices appear to be fully supported without having to mess around with DTHs. Also, the number of integrations is staggering.

I think this time it is for real. So I am saying farewell to this community as I will likely have little reason to visit. It is super easy to integrate ST into HA so I will keep the hub in case some devices integrate better or only with ST but I am not planning on adding back any Zwave/Zigbee devices. I hear crickets in my account…

There are too many users to thank for their contributions, and I am certain I cannot compile a complete list, but a few are: @JDRoberts @nathancu @jkp @MichaelS @johnconstantelo @orangebucket @rontalley @tonesto7 @krlaframboise

EDIT: …and I just found out you can tag max 10 people anyway…


If they introduce the 200 device limit on iOS I could be right behind you.


Sad but understandable.

At this point my personal hope for SmartThings (based on nothing but blind optimism) is that they will go all in on Matter, match or nearly match Apple Home for Matter feature compatibility (including voice navigation of the app), and thus provide a very strong Android base for both plug and play and power users. That would be good for everybody, much as Galaxy phones force iPhones to become better for consumers and vice versa. But who knows? If it does happen, we May all find ourselves back together in a year or two. :sunglasses:


I hate to see you go @aruffell , but I may not be too far behind. The last loophole for 200 devices for me is my old tablet on old Android and ST versions. I suspect that won’t last long. I know I need to add about 3 more devices very soon, so we’ll see how that goes.

Thank you for sharing those specs, that’s what I was looking for because I’m also seriously looking at Home Assistant.

So far everything I have will work with HA, but I just need to see how/if my Halo smoke alarms, water meter, and Aqara double rocker would work with HA.

If you can, share any other resource(s) you have for HA? Good luck!!!

1 Like

Sad to see you go @aruffell. It seems that one by one this community is running out of members. Honestly i understand that the 200 device limit is a bit excessive especially as it is not a number that is far out of reach for some.

I wish you all the best and hope that one day ST will become an actually compelling competitor to HA although i don’t think you will be switching back soon nomatter what changes as coming in the near future.

1 Like

Well, I’ve at least noticed that hardly anyone is saying they are moving on to Hubitat.

Just out of curiosity, was HE ever on your list of considerations, and how did you settle on HA?

1 Like

@mwav3 made that move a little while ago and posted quite a bit about it.


ST is likely growing furiously in members, but likely the ones that have a handful of devices, therefore the device hungry members are more of an issue than a benefit to ST. I can’t blame ST if this is the direction they are taking but that also means I can no longer be an ST user.

Also, with regards to the 200 limit… I can understand that there may be a resource issue when you go beyond a device threshold, but simply blocking new additions at 200, when a large number are cloud 2 cloud or even virtual switches seems to be a poor choice. I likely had 30 or more virtual switches… what load does that add to their system? Had I removed them, I could have added more physical devices adding to their burden. But here I am just guessing on their motivations…

1 Like

That’s because many already have! :wink: The Hubitat Community is full of exSmartThings, exWink, exIris, exVera, exHomeSeer, and exHomeAssistant folks…but mostly exSmartThings folks…

That being said, I do feel like Home Assistant has been coming on strong over the past 2 years. Great progress has been made to make it easier for ‘normal’ users.


Thanks…I get ya…

Anyway, those remaining seem to be leaning toward HA. So, that has me looking in that direction.

1 Like

I own the HE C5 hub and after moving a few devices to it, and then opting to integrate ST into HE I just left it at that. The biggest benefit I got from it was to keep using Echo Speaks via HE after ST killed Echo Speaks (GRRRRRR). I easily moved my webCore pistons sparking the notifications over to the HE instance, and having shared the ST devices over to HE, I was able to re-establish all the verbal notifications without having to natively add the devices to HE.

One issue I had was that most of the devices sparking the notifications were from a DSC (alarm) to ST integration that required an RPI running Alarmserver. I was weary of trying to move that integration over to HE. On a side note, the first day using HA, I found that adding 10 lines of configuration to one of its files allowed HA to pull in all my Alarm’s contacts without needing an additional RPI to integrate the two! So much simpler!

Going back to your question, HE was attractive for its local processing BUT the mobile app, and web GUI are very antiquated and not very appealing in my opinion. I also wondered what would happen once ST abandoned the IDE / Groovy… would HE do the same? Or would HE stay in the past? Also HE is a tiny company so while they have high flexibility and speed, they can also vanish quite quickly too. I wish them all the best, but I did not feel safe putting my eggs in their basket, so to speak.


Sorry to see you go Alex. Best of luck with your HA adventure.


The 200 device limit hit me hard last week as many already know. I was on the fence for a few days as I have a VERY complex install to migrate. After some very unhelpful interactions with ST support - I pulled the trigger and bought another Pi.

40 hours in and I don’t have all of the devices over yet - don’t get me started on the automations. Lets just say - I’m mid install. Did the initial install on a Pi48GB on SD but quickly decided the SSD upgrade was worth it. I have an SSD and that exact case on order right now. It should be here tomorrow.

If you aren’t computer literate and don’t understand command line ops - Home Assistant is easier than it was 5 years ago but its still not for you. But wow-the flexibility and the speed is about how much hardware I want to throw at it. Right now it’s running my lighting on botched Node Red flows with amazing speed!

Like JD - I’ll be hanging around helping ST newbies with core zwave stuffs… But I’ll be getting out of the webcore and autommations game as Webcore dies with the Groovy API and well TBH, I have little interest in the UI…

Oh - and - I wont be a troll. I urge all my peers who leave the platform to NOT BE A FREAKING TROLL.

Life is hard enough out there these days without someone beating you up because you have made a different decision. Like ST- -great! I hope it works. HE - good for you, I hope it gives you everything you want and local automations, HA , C4, Vera, Wink, or any other one I’ve missed? Great! Share what you know with the greater community. Just like when I ride my bike - I don’t berate others that don’t have a Harley. Life is better on two wheels no matter what kind of bike you ride. Same here in home automation.

Ride what you want. Just ride.


I don’t see any reason for Hubitat to leave Groovy. They don’t have a cloud-based IDE - the editor all runs on the hub. I think it’s a feature that if Hubitat went out of business today, my hub (and all my automations) would continue working mostly the same as it does now. But that’s also true of HA, which is a good choice, too.


Even if you are able to add the 3 devices, how long would it be feasible to keep exploiting a loophole to be able to service or grow your system? Unless ST reverses course, I would advise to start testing alternatives before you get stuck. Some of these alternatives can work along with ST so it would not be necessary to do a hard switch from one to the other, but if you do, plan it well. At first I pulled in all that I had on ST into HA (super easy via an API key) but then deleted the link as it was creating an awful mess with HA C2C integrations duplicating what was also integrated into ST. Alexa drove me insane telling me that I should not use the same name for more than one device… lol.

You can run HA on an RPI or a full blown server or some container on a NAS. I try to keep my systems separate so that one failure doesn’t tear down everything. To eliminate the need to administer the RPI OS, I went with HA’s RPI image that runs Home Assistant OS. The install was super easy… I flashed the regular Raspbian OS on the SD Card to update the RPI’s EEPROM firmware and to configure the RPI to boot from the SSD and I then flashed the HASS image to the SSD. Removed the SD Card and booted it all up. ‘Easy to boot’ :wink:

It is my understanding that if you ever want to upgrade, you can backup and restore on your new system. They also seem to have their “migration tool” issue figured out :wink:

The Zwave and Zigbee sticks I picked were mostly because I wanted them to be independent and of the latest generation. Both are supported even though HA doesn’t list the Z-Stick 7 yet, maybe because there is no series 700 benefit yet on HA. Also note that the Zigbee implementation does not yet support Zigbee 3.0 so your devices will be added using ZHA. This was not an issue with any of my devices.

Aqara integrations:

I read Halo went out of business while searching for an integration. If so it may be hard… I have yet to figure out a way to integrate my Nest Protect sensors… so I am in the same boat.


HomeKit. :wink: Only if you already have an iPhone, I don’t think it’s worth switching if you don’t. But if you do, the app is very good, the rules engine now supports quite complex rules, reliability is very good, Meross and Aqara both have low cost devices with excellent HomeKit integration. You will probably want to add a $99 HomePod Mini to allow out of home access and the $15 third party Home Plus app for a more complex rules engine, but the HomePod Mini also supports thread and everything will run locally except voice control, which is cool. It’s a very good plug and play solution.

If you’re more technical and you want to bring in Zwave devices, you’ll need to use a zwave hub and homebridge, which is doable but very much not plug and play.

Again, not for everybody, but certainly one for the candidate list if you already have an iPhone.

(Device limit is 150 devices per bridge, but no limit on the number of bridges, and a total limit of 1000 devices.)

Choice is good. :sunglasses:


I suspect that this is more like a line in the sand rather than an actuall issue with device load.

1 Like

I agree 100%. It is not easy and it will take some time for me to fully understand HA but I am already seeing the benefits so I am excited to learn and put in the effort.

Yeah… I don’t like the UI either. With so many devices I cannot call them “Light” so that the name is short enough to fit in the cards… Building the UI in HA is not super easy but OMG the stuff you can do has me salivating :wink:

I hope nothing I posted was seen as trolling. I am here to say good bye and share my excitement. ST was a good ride for the most part, and I learnt a lot, but I am now ready for a change.


Likely both. Funny thing is that when I reached the mythical 40 zwave devices on HA, I noticed some issues there too. There is surely a benefit to keeping the size of installs under a threshold from resources needed but also for things like the UI of the app. It was always apparent to me that it was designed for a much smaller number of devices. With the number of devices I had, it was so painful to use that I barely ever did mostly relying on automations or Alexa.

Are you using android or iOS?

1 Like